Network Event Impact Analysis via Segmented Diagnosis Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current network management technologies face challenges in efficiently analyzing and responding to complex network events, particularly in dynamic and heterogeneous environments like those found in modern data centers, where frequent changes and multiple concurrent faults require advanced diagnostic capabilities to maintain service level agreements (SLAs).
Innovation Solution
A programmable diagnosis model that uses element and service models to derive inference rules for impact analysis and root cause analysis (RCA), enabling the identification of cause-and-effect relationships and temporal dependencies across network resources, allowing for proactive and efficient management of network events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If distributed and fast diagnosis solution techniques are implemented to analyze dependent events in complex networks, then the ability to detect and respond to network failures improves, but the system complexity and computational overhead increase
Solution Approach 1:
The patent segments the network diagnosis problem into multiple layers: event collection layer, event correlation layer, and impact analysis layer. Each layer processes specific aspects of network events independently, reducing the complexity of any single component while maintaining comprehensive diagnostic capability across the entire system.
Solution Approach 2:
The system performs preliminary actions by pre-establishing event dependency models and correlation rules before failures occur. Event dependencies, impact relationships, and correlation logic are defined in advance, enabling rapid diagnosis when failures happen without requiring complex real-time computation of causal relationships.
2Measurement precision
If comprehensive event correlation and impact analysis are performed across all network resources, then the precision of root cause identification improves, but the processing time and computational resources required increase
Solution Approach 1:
Event dependency models, impact relationships, and correlation rules are pre-established before failures occur. The system pre-computes event dependencies and impact paths, so when a failure happens, the diagnosis process can quickly retrieve and apply these pre-defined relationships without performing complex real-time analysis of all network events.
Solution Approach 2:
The patent applies local quality by focusing correlation and impact analysis on specific event types and their relevant dependencies rather than uniformly analyzing all network events. Each event is processed with the specific correlation rules and impact relationships that are locally applicable to that event type, improving efficiency while maintaining accuracy.
3Reliability
If the system monitors and analyzes all network events in real-time to detect cascading failures, then the reliability of network management improves, but the energy consumption and processing load increase
Solution Approach 1:
The monitoring system is segmented into distributed event collectors that gather local events and a centralized correlation engine that processes dependencies. This segmentation allows energy-intensive correlation processing to be performed only on aggregated event data rather than raw individual events, reducing overall processing load and energy consumption while maintaining comprehensive monitoring capability.
Solution Approach 2:
The system performs partial action by focusing processing resources on correlated events and those with known dependencies rather than analyzing every single network event in equal detail. Events are processed with the level of analysis appropriate to their impact potential, reducing unnecessary processing energy consumption while maintaining reliability for critical events.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
Failure impact analysis (or "impact analysis") is a process that involves identifying effects of a network event that are may or will results from the network event. In one example, this disclosure describes a method that includes generating, by a control system managing a resource group, a resource graph that models resource and event dependencies between a plurality of resources within the resource group; detecting, by the control system, a first event affecting a first resource of the plurality of resources, wherein the first event is a network event; and identifying, by the control system and based on the dependencies modeled by the resource graph, a second resource that is expected to be affected by the first event.